Probabilistic and Reliability Analysis of an Intelligent Power Control for a Doubly Fed Induction Generator-Based Wind Turbine System.

The wind energy system's complicated characteristics have motivated many researchers to develop advanced and intelligent control strategies to ensure reliable and efficient system operation. Over the past few years, Artificial Neural Networks (ANN) have been widely employed in wind energy applications, and they are strongly recommended as a powerful and adequate tool for controlling wind turbine systems. The generator control block is a critical subsystem in the wind turbine chain, where a failed generator power control leads to an unstable operation process. The current work aims to verify the reliability of an ANN control of a DFIG integrated into a wind turbine system by considering the uncertainties in the machine parameters that affect the control's performance requirements. The major challenges encountered when applying reliability analyses are the system's high complexity and the expensive computational time required to achieve accurate results. In this context, the reliability assessment methodology adopted in this study combines Machine Learning techniques and advanced reliability approximation algorithms to optimize the computing time and ensure accurate results despite the system complexity. The obtained results demonstrate the reliability and effectiveness of ANN controllers in coping with machine parameter variations and maintaining the system at its optimal operating point. In addition, the results demonstrate the effectiveness of the reliability methodology implemented in this work by offering accurate approximations with a reasonable computing time. [ABSTRACT FROM AUTHOR]